AI-powered CLI coding agent with 20+ built-in tools, MCP support, and multi-model providers
This repository also configures its own agents. See what blade-code tells them →
Skill Claude Code needs its repo
A tool for converting Base64, a text format used to represent files or binary data, to and from ordinary text or files. It can also identify whether decoded content is text, JSON, an image, or a PDF.
Skill Claude Code
Review frontend code for best practices, bugs, and improvements. Use when reviewing PRs, checking code quality, or before committing.
Skill Claude Code
Explore and summarize popular GitHub projects. Use when the user wants to discover trending GitHub repositories, analyze project details, or get summaries of popular projects.
Skill Claude CodeCodex
An architectural guide to Blade Code, a TypeScript monorepo containing a coding agent, its tools, interfaces, services, and extension points.
Skill Claude CodeCodex
An orchestration layer for coding agents that manages decision loops, delegated agents, agent teams, goals, task lists, and scheduled work.
Skill Claude CodeCodex
A coordination system for teams of coding agents, with shared tasks, dependencies, messages, member status, and access checks.
Skill Claude CodeCodex
The main execution loop that moves a coding agent through model responses, tool calls, saved conversation state, error recovery, and completion checks.
Skill Claude CodeCodex
A system for persistent goals, task lists, independent task sessions, isolated worktrees, and cron, interval, or one-time schedules.
Skill Claude CodeCodex
A runtime for defining, running, pausing, resuming, and verifying child coding agents working on separate tasks.
Skill Claude CodeCodex
A project-wide quality and delivery workflow covering tests, release checks, builds, publishing, and bilingual documentation for Blade Code.
Skill Claude CodeCodex
A delivery system for building, testing, versioning, publishing, and documenting a TypeScript monorepo, which is one repository containing multiple related packages.
Skill Claude CodeCodex
A set of quality gates that checks performance, security, dependencies, test coverage, and stable output snapshots.
Skill Claude CodeCodex
A guide to tests that call real AI providers and verify complete user flows across the CLI, web interface, terminals, and ACP hosts.
Skill Claude CodeCodex
A testing guide for the CLI and web interfaces, covering unit, integration, end-to-end, and real-process test setup with shared fixtures and cleanup.
Skill Claude CodeCodex
A map of the system's extension ecosystem: MCP connections, plugins, skills, custom commands, hooks, and language servers, including how they enter an agent session.
Skill Claude Code
A codebase guide to hooks, which run around agent, tool, permission, session, and notification events to inspect or change behavior.
Skill Claude Code
A codebase guide to LSP, the standard protocol behind editor features such as symbol queries, file diagnostics, and language-aware code information.
Skill Claude CodeCodex
A codebase guide to MCP, a standard way for an agent to connect to external tools and content, including connections, authentication, dynamic tools, and recovery.
Skill Claude CodeCodex
A codebase guide to the full plugin lifecycle, from discovery and compatibility checks through installation, updates, disabling, and removal.
Skill Claude CodeCodex
A codebase guide to how skills and custom slash commands are discovered, loaded, parsed, and made available during an agent session.
Skill Claude CodeCodex
A guide to the CLI, terminal interface, web app, HTTP server, ACP hosts, and VS Code bridge that present the same agent session in different ways.
Skill Claude CodeCodex
A guide to ACP, a protocol that lets an IDE or other host communicate with an agent through sessions, files, terminals, permissions, and streamed updates.
Skill Claude CodeCodex
A codebase guide for the command-line startup path, including interactive, non-interactive, print, and structured JSONL output modes.
Skill Claude CodeCodex
A codebase guide for the server API that connects web clients and automation tools to sessions, tasks, live event streams, terminals, and browser runtimes.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: